# Sector8

- [Welcome to Sector8](https://docs.sector8.ai/guides/introduction.md): Pre-execution control for AI agents. Every proposed action returns ALLOW or DENY before it executes.
- [Onboarding](https://docs.sector8.ai/guides/onboard.md): One guided journey from published packages to ALLOW, DENY, evidence lookup, and observe/enforce proofs.
- [Onboarding assistant companion](https://docs.sector8.ai/guides/onboard-assistant.md): Setup-assistant instructions that follow the same onboarding journey and success criteria as guides/onboard.
- [Quickstart](https://docs.sector8.ai/guides/quickstart.md): Route your first proposed AI action through Sector8 and honor the returned decision.
- [Observe then enforce](https://docs.sector8.ai/guides/observe-then-enforce.md): Pilot path from evaluation-only observation to fail-closed enforcement, using verified SDK helpers.
- [Guard MCP Server](https://docs.sector8.ai/guides/mcp-server.md): Govern Claude Code tool calls routed through the sector8-guard MCP server with evaluate-before-dispatch decisions and fail-closed connector behavior.
- [Non-MCP integration](https://docs.sector8.ai/guides/non-mcp-integration.md): Wire custom or containerized tool runtimes through POST /api/v1/evaluate and honor ALLOW or DENY before dispatch.
- [Python SDK](https://docs.sector8.ai/guides/python-sdk.md): Use the Python SDK to evaluate proposed actions before dispatch, then optionally send telemetry.
- [TypeScript SDK](https://docs.sector8.ai/guides/typescript-sdk.md): Use the TypeScript SDK to evaluate proposed actions before dispatch, then optionally send telemetry.
- [Guardrails](https://docs.sector8.ai/guides/guardrails.md): How Sector8 evaluates proposed AI actions before execution and returns ALLOW or DENY.
- [Red teaming](https://docs.sector8.ai/guides/red-teaming.md): Test the live Sector8 contract: routed action, expected decision, evidence fields, and fail-closed behavior.
- [The Seven Invariants](https://docs.sector8.ai/guides/invariants.md): Foundational contracts for Sector8 evaluate-before-dispatch on routed integrations.
- [Compliance Overview](https://docs.sector8.ai/guides/compliance.md): How Sector8 provides pre-execution controls and decision evidence that support compliance workflows at the tool boundary.
- [EU AI Act](https://docs.sector8.ai/guides/eu-ai-act.md): How Sector8 can support EU AI Act-oriented controls and evidence at the AI tool execution boundary.
- [GDPR](https://docs.sector8.ai/guides/gdpr.md): How Sector8 supports GDPR-oriented controls with pre-execution evaluation and telemetry redaction evidence.
- [HIPAA](https://docs.sector8.ai/guides/hipaa.md): How Sector8 supports HIPAA-oriented PHI controls at the AI tool execution boundary.
- [API Overview](https://docs.sector8.ai/api-reference/overview.md): REST API mental model for Sector8 evaluate-before-dispatch decisions and supporting telemetry.
- [GET /api/health](https://docs.sector8.ai/api-reference/health.md): Returns the current health and capability state of the Sector8 gate.
- [POST /api/v1/evaluate](https://docs.sector8.ai/api-reference/evaluate.md): Evaluate a proposed tool or agent action before dispatch. Returns ALLOW or DENY with an evidence record.
- [POST /api/v1/telemetry](https://docs.sector8.ai/api-reference/analyze.md): Ingest a completed LLM interaction for telemetry, compliance processing, and post-execution analysis.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.